Hah, my university (which for the record was Durham, England) was so incompetent that they attempted to make students sign an agreement handing over rights to code after we started the course. It was just some form they handed out in a lecture early on in the first year and required us to sign. Rather unbelievably it not only covered stuff we did for assignments but anything we wrote whilst enrolled on the course!
I am forgetting the details, but I think it also said something to the effect of any patentable inventions we created were also owned by the university, and they gave us a whole spiel about students who formed a "partnership" with the university. Their carrot was that the university would lend its supposedly solid and respectable name to the business, and let the student use their own inventions (!), and in return the university got some percentage of the profits.
I signed (because there was no alternative) and then checked with some other people, who all confirmed that it was unenforceable BS. They can't change the terms of a contract after you started the course even if you do sign a piece of paper. I figured if they were that incompetent then they wouldn't find out about any code I wrote whilst I was there. And I was right.
K&R, and the linux/GNU write well, wereas their MS counterparts wirte pretty crappy stuff. I would also venture to guess that the code alone can serve as an example of how to write code.
You're kidding, right? We're talking about the kernel that doesn't bother with unit tests or actual release cycles, and which thinks that "our users find our bugs" is a good replacement for a QA process?
Look, I like Linux and use it a lot, but to say that they "write well" and Microsoft "write pretty crappy stuff" is reveal a limited exposure to professional software development.
BTW I disagree there's much to be learned from looking at old codebases. It's like saying modern engineers can learn something from examining rusty old steam engines. Sure, it's useful to know where it came from, but that can be covered quite well by a book, you don't need to spend time actually reading the old code. Because nearly without question the worst code I see in operating systems is always the oldest.
My point is exactly that you are not guaranteed the system will work. The people patching the software for a distribution usually have a lesser understanding of the software than the authors do. This can and does lead to critical bugs being added in the process of distribution.
Actually this is pretty difficult. It's technically possible if you pay close attention to the versioning used by the distributor and ensure your packages will always override theirs. It's a lot of effort and really not user friendly or well supported. And it's not an app store, which was my point.
Well, right, I think we're saying the same thing and agreeing:) The core of the problem was Debians requirement that they should be able to ship unapproved patches to Firefox and still call it Firefox. That makes no sense for any program that is trying to build a brand reputation. That's why I said an open source "app store" would have a very different philosophy to an open source "repository" where the understanding is that you can "apt-get install firefox" and possibly get something that is not what the Firefox developers actually created.
Well, there are a bunch of features mobile app stores have that repositories do not. For instance, the ability to post comments, or to rate programs.
The biggest difference though is in philosophy. Even if you assume a completely authoritarian app store like the iPhones, the apps you download from it are basically what the developers created. You are getting it "as the creator intended". And if you are comparing against Android rather than the iPhone, then the app store is very open, you can upload almost whatever you want within some basic limitations, like, you can't upload an app that violates some other services terms-of-service. Also, an app store is always fresh, because the latency from developers finishing QA on their binary and it being available to download is seconds, at least for Android.
In contrast, Linux distro repositories have a different philosophy:
They reserve the right not only to reject your software for any reason, but also to modify it via patches as well. The user is not informed that any patching has taken place. Sometimes this patching improves the software, but sometimes it totally breaks it. There are many examples of this (eg openssl).
This silent tampering is extensive and distributors are loathe to give it up. When Mozilla decided they didn't want the Firefox brand tarnished by extensive Debian patches, Debian decided they'd rather rename the product "Iceweasel" than give up this control.
Distros are not fresh. Typically the software that was around at the time of release is frozen and updates from upstream are not made available, unless they are security updates. Even then some distros prefer to "backport" security fixes, rather than simply follow upstream versioning. This results in a steady stream of useless bug reports to upstream for problems that were long since fixed. Once again, the developers are not in control of their own software.
If Novell are actually interested in the app store approach, they're going to have to convince the suse developers to give up that level of control and make automated import of upstream binaries the norm. No more "packagers" for applications - that role will have to be obsoleted. And then they'll have to convince upstream developers to actually submit those binaries.
I am doubtful that this will happen. Some years ago I promoted a more normal approach to app distribution on Linux (not an app store, but true web-based distribution). I was flatly told by several distribution employees that they weren't interested in losing control of the total software experience like that, and there would be no change in policy whilst they were around. So I gave up. These days I focus on Android - it's actually got a sane design and software distribution mechanism. Many of the things I wanted to see in Linux are in Android. Novell should be looking at how they can get in on that..... unless they still think Linux is a viable mass-market desktop?
I'm not sure why you'd need laws. Presumably, game publishers don't want to allow internet resale because the ease of finding people to sell to online is huge and if there was no physical getting off the couch/posting a disc nearly everyone would do it. I'm guessing that this would require the games to get more expensive (or lower budget) to compensate.
If there was big demand and people really cared about this issue the free market would take care of it, because games that could be resold online would be more popular than those which couldn't, and thus more profitable despite being more expensive. I don't see that though, so I'm guessing the real answer is that most people don't care about it enough to let it dictate their choice of games. For instance I don't care about selling my old games much. I'll happily take the ease of internet distribution over that any day.
Hmm. Sounds like the real problem here is that autopilots are not built to explain their decisions? I mean, what if there was a reason for the auto-pilot to be doing those things that made sense, you just didn't know what they were? Wasn't one of the air disasters mentioned earlier in the discussion where the auto-pilot dived to maintain speed after there was ice on the plane, and the pilot overrode it because he didn't understand why?
I don't think speculation is pointless. We have one extremely unlikely possibility, and one extremely likely possibility. Saying speculation is pointless implies they are equally likely.
Hm, right, cuz nobody ever used an easily guessable password before. This is going to be the tool of choice for spurned partners everywhere.... that said, it's still a good idea. I'm just skeptical a password is going to cut it.
Do you realize how paranoid you sound? Why don't we apply Occams Razor here..... let's see, either the anti-piracy agency as you call it could be playing some subtle mind game in an attempt to discredit, uh, people who are already criminals and have a history of stupid harassment. Or it could be yet another inspired piece of retardedness from the same people who brought you "let's DoS a law firm with fraudulent wire transfers". Hmmm, I wonder which is more likely?
What concerns exactly? That it's illegal to download copyrighted stuff? That's not a concern that will be taken seriously by anyone outside the pirate party. I mean this is the party who seriously proposes replacing pharma patents with all drug R&D being government funded. Their policies appear to be incredibly flimsy, there's not even any discussion of the content providers POV on their English website. It's just "p2p should be free, drm should be illegal, nobody needs to make money after 5 years anyway".
The G1 browser is certainly less pretty than the iPhone/Pre browsers. And thanks to Apple it doesn't have multi-touch zooming (I guess Palm are confident they have enough mobile patents to cross license). However, one thing I've noticed is that drag response is way better on the G1 than the iPhone and apparently the Pre. I'm not sure what it is, but the checkerboard pattern is something I see a lot of on iPhones. I did a side by side comparison of my G1 and a friends iPhone, and the dragging response wasn't even comparable, the G1 blew it away. I'd love to know why, given that they are based on the same code. Pre seems to have the same problem!
Have you ever actually written a Java program? It's got the name "write once, debug everywhere" for a reason. And always backwards compatible? Huh? Is that why real Java apps usually come with a complete, private copy of the JVM?
If you believe the judge was truly biased (and not simply being an expert in his field) then you should side with no side, because the question of guilt is still open. Siding with the Pirate Bay because they claim the judge is biased makes no sense... the judge may or may not have been biased by his affiliations, but the Pirate Bay are definitely biased.
Eh what? Lost is the most extreme case of forward planning I've ever seen. If you think they don't have it all planned out you're not paying attention. Did you notice that the 4 toed statue was first encountered in passing in season 2, then barely featured again until the last episodes of season 5? How about the way Pierre Chang first appeared with a prosthetic right arm in a mysterious video way back at the start of season 2, and right at the end of season 5 you see the accident in which he got that bad arm? He wasn't just thrown in randomly as "mysterious dude with bad arm" and then reintroduced later, it hangs together too well for that.
Using the GPL wouldn't help in this case. Even if the dude had provided everyone who bought the app with the source, what will they do with it? The source isn't useful unless you're in the developer program. And you'd only get it when you buy the app anyway.
Eh? But, game engines run in "kernel mode" on games consoles all the time... and they're often the same codebase these days. I agree that it would not make much of a performance difference though.
Generally, in order of preference I'd look at D, C# and then Java as what Objective-C should aspire to. But it'd have to change so radically why bother? If Apple insist on an obscure language with poor toolchain support D at least has the advantage of being a really well designed language, with lots of useful features.
Well, not much, but then what do you have to gain by using native apps on Linux? Isn't the real problem here that Linux has no real competitive edge over Windows? It's not like GTK+ or Qt apps have some incredible edge over Windows apps that can make up for lost functionality.
I thought the business model for Canonical was enterprise desktop Linux? I mean, they have to grow up and become a profitable firm at some point right, even Shuttleworth will run out of money one day. But if their marketing for Ubuntu is "it's good because it DOESN'T run your existing line-of-business applications" I don't think they're going to get far. Canonical is a toy company, still, and always has been.
Slashdot Mirror
Hah, my university (which for the record was Durham, England) was so incompetent that they attempted to make students sign an agreement handing over rights to code after we started the course. It was just some form they handed out in a lecture early on in the first year and required us to sign. Rather unbelievably it not only covered stuff we did for assignments but anything we wrote whilst enrolled on the course!
I am forgetting the details, but I think it also said something to the effect of any patentable inventions we created were also owned by the university, and they gave us a whole spiel about students who formed a "partnership" with the university. Their carrot was that the university would lend its supposedly solid and respectable name to the business, and let the student use their own inventions (!), and in return the university got some percentage of the profits.
I signed (because there was no alternative) and then checked with some other people, who all confirmed that it was unenforceable BS. They can't change the terms of a contract after you started the course even if you do sign a piece of paper. I figured if they were that incompetent then they wouldn't find out about any code I wrote whilst I was there. And I was right.
You're kidding, right? We're talking about the kernel that doesn't bother with unit tests or actual release cycles, and which thinks that "our users find our bugs" is a good replacement for a QA process?
Look, I like Linux and use it a lot, but to say that they "write well" and Microsoft "write pretty crappy stuff" is reveal a limited exposure to professional software development.
BTW I disagree there's much to be learned from looking at old codebases. It's like saying modern engineers can learn something from examining rusty old steam engines. Sure, it's useful to know where it came from, but that can be covered quite well by a book, you don't need to spend time actually reading the old code. Because nearly without question the worst code I see in operating systems is always the oldest.
My point is exactly that you are not guaranteed the system will work. The people patching the software for a distribution usually have a lesser understanding of the software than the authors do. This can and does lead to critical bugs being added in the process of distribution.
Actually this is pretty difficult. It's technically possible if you pay close attention to the versioning used by the distributor and ensure your packages will always override theirs. It's a lot of effort and really not user friendly or well supported. And it's not an app store, which was my point.
Well, right, I think we're saying the same thing and agreeing :) The core of the problem was Debians requirement that they should be able to ship unapproved patches to Firefox and still call it Firefox. That makes no sense for any program that is trying to build a brand reputation. That's why I said an open source "app store" would have a very different philosophy to an open source "repository" where the understanding is that you can "apt-get install firefox" and possibly get something that is not what the Firefox developers actually created.
Well, there are a bunch of features mobile app stores have that repositories do not. For instance, the ability to post comments, or to rate programs.
The biggest difference though is in philosophy. Even if you assume a completely authoritarian app store like the iPhones, the apps you download from it are basically what the developers created. You are getting it "as the creator intended". And if you are comparing against Android rather than the iPhone, then the app store is very open, you can upload almost whatever you want within some basic limitations, like, you can't upload an app that violates some other services terms-of-service. Also, an app store is always fresh, because the latency from developers finishing QA on their binary and it being available to download is seconds, at least for Android.
In contrast, Linux distro repositories have a different philosophy:
They reserve the right not only to reject your software for any reason, but also to modify it via patches as well. The user is not informed that any patching has taken place. Sometimes this patching improves the software, but sometimes it totally breaks it. There are many examples of this (eg openssl).
This silent tampering is extensive and distributors are loathe to give it up. When Mozilla decided they didn't want the Firefox brand tarnished by extensive Debian patches, Debian decided they'd rather rename the product "Iceweasel" than give up this control.
Distros are not fresh. Typically the software that was around at the time of release is frozen and updates from upstream are not made available, unless they are security updates. Even then some distros prefer to "backport" security fixes, rather than simply follow upstream versioning. This results in a steady stream of useless bug reports to upstream for problems that were long since fixed. Once again, the developers are not in control of their own software.
If Novell are actually interested in the app store approach, they're going to have to convince the suse developers to give up that level of control and make automated import of upstream binaries the norm. No more "packagers" for applications - that role will have to be obsoleted. And then they'll have to convince upstream developers to actually submit those binaries.
I am doubtful that this will happen. Some years ago I promoted a more normal approach to app distribution on Linux (not an app store, but true web-based distribution). I was flatly told by several distribution employees that they weren't interested in losing control of the total software experience like that, and there would be no change in policy whilst they were around. So I gave up. These days I focus on Android - it's actually got a sane design and software distribution mechanism. Many of the things I wanted to see in Linux are in Android. Novell should be looking at how they can get in on that ..... unless they still think Linux is a viable mass-market desktop?
I'm not sure why you'd need laws. Presumably, game publishers don't want to allow internet resale because the ease of finding people to sell to online is huge and if there was no physical getting off the couch/posting a disc nearly everyone would do it. I'm guessing that this would require the games to get more expensive (or lower budget) to compensate.
If there was big demand and people really cared about this issue the free market would take care of it, because games that could be resold online would be more popular than those which couldn't, and thus more profitable despite being more expensive. I don't see that though, so I'm guessing the real answer is that most people don't care about it enough to let it dictate their choice of games. For instance I don't care about selling my old games much. I'll happily take the ease of internet distribution over that any day.
Hmm. Sounds like the real problem here is that autopilots are not built to explain their decisions? I mean, what if there was a reason for the auto-pilot to be doing those things that made sense, you just didn't know what they were? Wasn't one of the air disasters mentioned earlier in the discussion where the auto-pilot dived to maintain speed after there was ice on the plane, and the pilot overrode it because he didn't understand why?
10% profitability is roughly the average across all industries, actually.
I don't think speculation is pointless. We have one extremely unlikely possibility, and one extremely likely possibility. Saying speculation is pointless implies they are equally likely.
Hm, right, cuz nobody ever used an easily guessable password before. This is going to be the tool of choice for spurned partners everywhere .... that said, it's still a good idea. I'm just skeptical a password is going to cut it.
Do you realize how paranoid you sound? Why don't we apply Occams Razor here ..... let's see, either the anti-piracy agency as you call it could be playing some subtle mind game in an attempt to discredit, uh, people who are already criminals and have a history of stupid harassment. Or it could be yet another inspired piece of retardedness from the same people who brought you "let's DoS a law firm with fraudulent wire transfers". Hmmm, I wonder which is more likely?
What concerns exactly? That it's illegal to download copyrighted stuff? That's not a concern that will be taken seriously by anyone outside the pirate party. I mean this is the party who seriously proposes replacing pharma patents with all drug R&D being government funded. Their policies appear to be incredibly flimsy, there's not even any discussion of the content providers POV on their English website. It's just "p2p should be free, drm should be illegal, nobody needs to make money after 5 years anyway".
The G1 browser is certainly less pretty than the iPhone/Pre browsers. And thanks to Apple it doesn't have multi-touch zooming (I guess Palm are confident they have enough mobile patents to cross license). However, one thing I've noticed is that drag response is way better on the G1 than the iPhone and apparently the Pre. I'm not sure what it is, but the checkerboard pattern is something I see a lot of on iPhones. I did a side by side comparison of my G1 and a friends iPhone, and the dragging response wasn't even comparable, the G1 blew it away. I'd love to know why, given that they are based on the same code. Pre seems to have the same problem!
The 1.5 cupcake update added that exact ability. It's rolling out now.
Ah ha, the business model behind Android finally reveals itself :)
Have you ever actually written a Java program? It's got the name "write once, debug everywhere" for a reason. And always backwards compatible? Huh? Is that why real Java apps usually come with a complete, private copy of the JVM?
If you believe the judge was truly biased (and not simply being an expert in his field) then you should side with no side, because the question of guilt is still open. Siding with the Pirate Bay because they claim the judge is biased makes no sense ... the judge may or may not have been biased by his affiliations, but the Pirate Bay are definitely biased.
Eh what? Lost is the most extreme case of forward planning I've ever seen. If you think they don't have it all planned out you're not paying attention. Did you notice that the 4 toed statue was first encountered in passing in season 2, then barely featured again until the last episodes of season 5? How about the way Pierre Chang first appeared with a prosthetic right arm in a mysterious video way back at the start of season 2, and right at the end of season 5 you see the accident in which he got that bad arm? He wasn't just thrown in randomly as "mysterious dude with bad arm" and then reintroduced later, it hangs together too well for that.
Interesting that you chose "half". I believe Microsoft have claimed that the better anti-piracy code in Vista halved their piracy rate.
Using the GPL wouldn't help in this case. Even if the dude had provided everyone who bought the app with the source, what will they do with it? The source isn't useful unless you're in the developer program. And you'd only get it when you buy the app anyway.
Eh? But, game engines run in "kernel mode" on games consoles all the time ... and they're often the same codebase these days. I agree that it would not make much of a performance difference though.
Generally, in order of preference I'd look at D, C# and then Java as what Objective-C should aspire to. But it'd have to change so radically why bother? If Apple insist on an obscure language with poor toolchain support D at least has the advantage of being a really well designed language, with lots of useful features.
Well, not much, but then what do you have to gain by using native apps on Linux? Isn't the real problem here that Linux has no real competitive edge over Windows? It's not like GTK+ or Qt apps have some incredible edge over Windows apps that can make up for lost functionality.
I thought the business model for Canonical was enterprise desktop Linux? I mean, they have to grow up and become a profitable firm at some point right, even Shuttleworth will run out of money one day. But if their marketing for Ubuntu is "it's good because it DOESN'T run your existing line-of-business applications" I don't think they're going to get far. Canonical is a toy company, still, and always has been.